, Volume 37, Issue 1-2, pp 35-51

Multiproxy summer and winter surface air temperature field reconstructions for southern South America covering the past centuries

Purchase on Springer.com

$39.95 / €34.95 / £29.95*

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access

Abstract

We statistically reconstruct austral summer (winter) surface air temperature fields back to ad 900 (1706) using 22 (20) annually resolved predictors from natural and human archives from southern South America (SSA). This represents the first regional-scale climate field reconstruction for parts of the Southern Hemisphere at this high temporal resolution. We apply three different reconstruction techniques: multivariate principal component regression, composite plus scaling, and regularized expectation maximization. There is generally good agreement between the results of the three methods on interannual and decadal timescales. The field reconstructions allow us to describe differences and similarities in the temperature evolution of different sub-regions of SSA. The reconstructed SSA mean summer temperatures between 900 and 1350 are mostly above the 1901–1995 climatology. After 1350, we reconstruct a sharp transition to colder conditions, which last until approximately 1700. The summers in the eighteenth century are relatively warm with a subsequent cold relapse peaking around 1850. In the twentieth century, summer temperatures reach conditions similar to earlier warm periods. The winter temperatures in the eighteenth and nineteenth centuries were mostly below the twentieth century average. The uncertainties of our reconstructions are generally largest in the eastern lowlands of SSA, where the coverage with proxy data is poorest. Verifications with independent summer temperature proxies and instrumental measurements suggest that the interannual and multi-decadal variations of SSA temperatures are well captured by our reconstructions. This new dataset can be used for data/model comparison and data assimilation as well as for detection and attribution studies at sub-continental scales.